Claude Fable 5.2 Price and Limits: Is This Week’s Reset Offer Worth Switching?
Anthropic launched Claude Fable 5.2 and Mythos 5.2 on November 9, 2026. API list price dropped to $8 input / $40 output per million tokens. This week only, eligible paid plans get extra Fable resets—and Pro/Team standard seats get Fable inside weekly limits up to 50%. Here is what that means for quotas, and when to stay on Opus 5.5.
bufferAbout the author8 min read
Anthropic released Claude Fable 5.2 and Mythos 5.2 on November 9, 2026. The useful part for anyone watching quotas: the API list price is lower than Fable 5.1, there is an optional fast mode, and this week only eligible paid plans get extra Fable usage. This page sticks to official numbers. For the older comparison table that still lists Fable 5.1 at $10/$50, see our LLM API pricing comparison. For how weekly and five-hour caps work on Pro/Max/Ultra, see Claude usage limits explained.
API price: Fable 5.2 vs what our price page still says
Official list prices from Anthropic’s Claude Fable product page and the Claude Platform pricing docs (checked November 10, 2026, Beijing time):
| Model / mode | Input / MTok | Output / MTok | Cache hits / MTok |
|---|---|---|---|
| Fable 5.2 (standard) | $8 | $40 | $0.20 |
| Fable 5.1 (previous) | $10 | $50 | $0.25 |
| Fable 5.2 fast mode | $16 | $80 | (cache multipliers stack on fast rates) |
| Mythos 5.2 (same rates; limited availability) | $8 | $40 | $0.20 |
| Opus 5.5 (Anthropic’s recommended daily driver) | $4 | $20 | $0.20 |
What that means in plain words:
- Fable 5.2 is 20% cheaper on both input and output than Fable 5.1 ($8/$40 vs $10/$50).
- Fast mode is 2× the standard Fable 5.2 rate ($16/$80) for up to 2.5× faster token generation.
- Our site price table was written against Fable 5.1. Until that page is updated, treat $8/$40 as the current Fable 5.2 list price and $10/$50 as the previous generation.
Other official pricing notes worth keeping:
- Cache writes: 5-minute write $10 / MTok, 1-hour write $16 / MTok on Fable 5.2 (docs table).
- Batch API: Fable 5.2 batch is $4 input / $20 output per MTok (50% off).
- US-only inference (
inference_geo: "us") is 1.1× on Claude 4.6+ models, including Fable 5.2.
API model id for Fable 5.2: claude-fable-5-2.
This week’s launch promotion (time-boxed)
Source: Anthropic’s launch post plus the Help Center article Claude Fable 5.2 launch offers. Mark these as this-week / launch-window only—do not treat them as permanent plan rules.
Who is eligible
- Available on Pro, Max, Ultra, standard and premium Team seats, and premium seats on legacy seat-based Enterprise.
- Not on Free; not on standard seats on legacy seat-based Enterprise; not on self-serve Enterprise; not on usage-based Enterprise (per launch offers).
- HIPAA-ready Enterprise is covered separately in the FAQ: no Fable limit resets.
- Resets only go to accounts with paid billing in place before November 9, 2026 at 10:00 am PT (Beijing time Nov 10, 02:00). Team/Enterprise: the organization’s billing must have been in place by then; seat holders must also have joined by that cutoff. If the subscription started before the cutoff but was still in a free trial, resets arrive once it converts to paid (self-serve Team trials count as paid).
Offer A — Pro plans and Team standard seats
From launch on November 9, 2026 through November 16, 2026 at 11:59 pm PT (Beijing time Nov 17, 15:59):
- All Fable models (5.2, 5.1, and 5) are included in your plan’s weekly usage limits.
- Cap: up to 50% of weekly usage limits on Fable.
- Plus exactly one Fable limit reset—not the 1–3 tenure ladder (that ladder is only for Max / Ultra / Team premium / legacy Enterprise premium). Same billing cutoff applies. That reset expires November 16, 2026 at 11:59 pm PT (Beijing time Nov 17, 15:59).
- On Pro and Team standard seats, once you use up that 50% Fable allowance or the week ends (November 16, 2026 at 11:59 pm PT / Beijing time Nov 17, 15:59), further Fable usage is billed to usage credits if you have them turned on—it no longer draws from weekly limits after the week ends. Unused 50% allowance does not carry over, and the unused single reset expires with the window (per launch offers).
Offer B — Max, Ultra, Team premium, legacy Enterprise premium
- One to three Fable limit resets based on subscription tenure (measured from first paid bill; Team/Enterprise from the organization’s first paid bill):
| Subscription length | Fable resets |
|---|---|
| Less than six months | One |
| Six to 12 months | Two |
| More than 12 months | Three |
- These resets expire March 8, 2027 at 11:59 pm PT (Beijing time March 9, 2027, 15:59).
What a Fable limit reset actually does
Per the Help Center:
- Weekly limit: restores usage spent on Fable models only. Usage on other models stays.
- Five-hour window: refills the whole five-hour window, including usage spent on other models.
- It does not raise your limits. The 50% Fable weekly cap still applies where that rule exists.
- Use it in Settings > Usage → Limit resets → Reset for free. You can use a reset before you hit the wall; once used, it cannot be undone. Cancel/downgrade before using it and it goes away.
Verified Ultra accounts without a five-hour limit: a reset only affects the weekly limit.
When to switch vs stay on Opus 5.5
Anthropic’s launch post is blunt: for most coding and knowledge work, start with Claude Opus 5.5, then move up to Fable/Mythos only when the job is open-ended and hard enough. Fable 5.2’s cyber and biology safeguards are as robust as Fable 5.1’s but produce fewer unnecessary blocks: early testing across claude.ai, Claude Code, and Cowork saw about 55% fewer blocks vs Fable 5.1—about 60% fewer in cybersecurity and about 40% fewer in biology (per the launch post). Blocked cyber queries fall back to Opus 4.8; biology to Opus 5—not “zero blocks.”
Practical cut:
- Stay on Opus 5.5 for day-to-day Claude Code / chat if your weekly bar is already tight.
- Switch to Fable 5.2 for multi-hour migrations, ambiguous research, or agent orchestration—especially this week if you are on Pro/Team standard and the included 50% allowance is still open.
- Fast mode is a latency buy, not a capability upgrade: you pay 2× list for up to 2.5× speed on Claude Code, Claude Platform, Google Cloud Claude models, and Microsoft Foundry (Anthropic hosted). Mythos fast mode is only for Verification Program orgs.
- Mythos 5.2 is the same underlying model without the Fable cyber/biology safeguards. Access is only via the Cyber Verification Program or the Life Sciences Verification Program—not a self-serve toggle.
Other launch notes that affect integrations: outputs up to 30% faster than 5.1; preserved thinking now applies to all accounts (API feature against illicit distillation).
Claude Fable 5.2 is now available in Cursor. On Cursor’s own benchmark, CursorBench, Cursor reports 60% at max effort—which it calls the top model on that benchmark—and 56.9% at default effort, up from 49.2% for Fable 5.1 (see Cursor’s post, Nov 9, 2026 UTC). Max effort uses more quota; the default setting already beats 5.1 by a good margin.
FrontierCode 2.0: half-cost match is scoped to one benchmark
On Anthropic’s announcement FrontierCode 2.0 section: in early testing, only on the FrontierCode 2.0 test, Fable 5.2 at High effort roughly matches Opus 5.5 at Max for about half the cost per task (the prose cites 45.3% vs 45.5%). The same page lists 49.1% for Fable 5.2 at Xhigh effort in the benchmarks table.
Read that only as a FrontierCode 2.0 benchmark result—not as overall parity with Opus 5.5. Anthropic still recommends Opus 5.5 as the daily driver. Other charts on the same page (WANDR, HLE, Chartography) have their own cost curves and must not inherit this half-cost wording.
Mainland China availability
Not verified for mainland China. Claude is not officially available for consumer use in mainland China. This article does not describe workarounds, relays, or third-party keys. If your org has a documented enterprise path, use Anthropic’s own channels.
FAQ
Is Fable 5.2 cheaper than Opus 5.5?
No on list price. Opus 5.5 is $4/$20; Fable 5.2 is $8/$40. Fable is the expensive specialist. The win vs last generation is vs Fable 5.1 ($10/$50), not vs Opus. “Roughly half the cost” appears only in Anthropic’s FrontierCode 2.0 High-vs-Max note—see the scoped section above.
Does the 50% Fable allowance on Pro last forever?
No. For Pro and Team standard seats it is a one-week launch window ending November 16, 2026 at 11:59 pm PT (Beijing time Nov 17, 15:59), then Fable returns to usage credits.
Where do Max/Ultra sit after the promo?
Per the Fable 5.2 launch offers help page, Fable usage is capped at 50% of weekly usage limits on plans/seats that include Fable (including Max/Ultra). The launch gift for Max/Ultra is the extra tenure-based resets (valid through March 8, 2027 at 11:59 pm PT / Beijing time March 9, 2027, 15:59), not a brand-new 50% rule. For plan comparisons, see also Claude Pro vs Max.
How does this connect to Claude Code quotas?
Subscription caps are still the five-hour + weekly system described in usage limits. A Fable reset can refill the five-hour window entirely, but the weekly Fable share only restores Fable spend.